Lady Anne Boleyn Rev07 Testing

[larry cottrill]

The stainless Lady Anne Boleyn Rev07 prototype was first tested last evening, 26 July 2007, fueling with propane vapor at 12 PSIG and using 30 PSIG starting air. I should have video ready next Monday (probably via YouTube). Again, this is the first Lady Anne variation that I've actually had sustaining (though there have been earlier successes by others).

The good news is: she starts right up and runs.

The bad news is: she heats right up and quits.

Basically, the engine as built is slightly out of tune for hot running. Starting was achieved with 30 PSI compressed air in the starting air pipe. Pulsing the starting air did NOT help - a steady stream of air allowed the engine to start as soon as the fuel valve was sufficiently opened. Just turn on the spark, press the air button, start opening up the fuel valve and the engine is running perfectly within seconds. You can shut off the air immediately and she just keeps going - no "easing off" of starting air needed at all. Once again, the properly aimed starting air tube is fully vindicated.

However, once the engine heats up enough for the pulse frequency to rise somewhere between a half-step and a full-step above its initial tone, the engine abruptly flames out. At that point, the mid cone and choke cone are well heated (clearly glowing bright orange as seen in "dusk" light). Running is smooth right up to flameout.

While disappointing, this result is certainly not mysterious, nor totally unexpected in a "new" design. My hypothesis is as follows: The main pipe of the engine is "shortened" acoustically by gradual heating (just another way of saying that the frequency rises as the speed of sound in the pipe increases with average temperature rise), while the intake temperature is almost perfectly stable (because its contents are entirely purged and refreshed with outside air in each cycle). The main pipe frequency becomes a bit too high to maintain resonance. This is possible even when the overall Q of the intake is fairly low (as it usually is with my designs).

If this is the correct explanation, the most desirable final solution would be to shorten the intake slightly. To test the hypothesis, and get an idea of how much to modify it, lengthening the tail cone with a short cylinder should do it. Unfortunately, my four little thrash mount lugs get in the way of doing this with a snug-fitting sliding sleeve. Probably, the thing to do is to verify that the plain steel version behaves the same way and mess with it to get good running (since it is MUCH easier for me to tweak experimentally), then apply the lessons learned to the stainless prototype.

As will be seen in the video (next week), I also tried running with radically reduced fuel pressure (about 5 PSIG). Amazingly, the engine was able to run with this, although starting was pretty tricky. With the lowered fuel flow this provided, the Lady was able to run much longer before flaming out, but of course, this was just the result of much slower heating of the engine shell. It occurs to me that none of the runs I got were really very high power level operation and the last run was really low, which might indicate at least some reasonable range of throttleability. My original guess was that flameout was from rich running, but I think it became clear later that this was erroneous. It is possible that the 3/16-inch fuel pipe used is too restrictive for full power from such a low pressure regulated supply.

Summing up the basic observations:
- The engine starts up and sustains quickly and easily
- The engine becomes quickly de-tuned as temperature rises
- My guess at 30 PSIG starting air was almost spot on for good starting
- Engine heating pattern was exactly as expected from earlier work by others
- Heating of the chamber "problem area" just aft of the intake port did not seem excessive
- The engine was always running at fairly low levels, with no visible flame from intake or tail cone
- Engine mounting scheme seems to work well

L Cottrill

[Irvine.J]

Great work larry, and congrats on your engine.
I'm very happy its sustaining well now, I remember your song now too "Our wheels turn slow but we get there" or something

Thats one heck of a short intake already Larry! You said its about an inch D right on that model (if memory serves), if you do slide in an intake (1mm wall thickness being -2mm overall D) try it at the same length before cutting the intake flare off to make it shorter. The increased resistance might just be all she needs. I found from the large intakes on the earlier models simply a reduction in pipe diameter really helped reign them in. If its not the 1 inch D like we discussed forget about all that.

This flame out during temperature increase is very very interesting to say the least, and you say that your intake temperature is perfectly stable, what if I put this to you...

Since the intake is drawing big breaths across that steadily increasing surface temp, might that have been something we completely brushed over but should have considered a little more closely. Is it possible its having a slightly increased effect having the intake so close, what do you think about that, I really can't be sure?
Anyway thats a bit of excited speculation from my part, just wondering how you felt about that idea?

Good stuff larry, I bet you jumped for joy when she fired up that fast and easily! You know she is a real beauty, your skill on these matters has always been incredible, and I'm sure you'll have that whopping great blue aura out the tailpipe real soon, keep up the good work, and congrats.
Now, you'd better go catch your horses because at the frequency these things run at I bet they scattered like crazy

[Eric]

Also having the intake close to the tail pipe will reduce its effective area by reducing the volume of air the engine can breath, as well as sucking hot air in at a lower density.

Can you actively cool it and see what happens?

Eric

[larry cottrill]

Great work larry, and congrats on your engine.

Thank you, sir!

Thats one heck of a short intake already Larry! You said its about an inch D right on that model (if memory serves), if you do slide in an intake (1mm wall thickness being -2mm overall D) try it at the same length before cutting the intake flare off to make it shorter. The increased resistance might just be all she needs. I found from the large intakes on the earlier models simply a reduction in pipe diameter really helped reign them in. If its not the 1 inch D like we discussed forget about all that.

This intake is a thick-walled "dairy pipe" (sanitary tubing) section - exactly 1 inch OD and .062 inch thickness. That translates to an OD of 25.4 mm with an ID of 0.876 inch or 22.25 mm - so, it's not as big and short as you might think at first glance. The intake on the steel engine is a little smaller, maybe 20 mm or something (it's nominally 3/4-inch EMT, but the real ID is always somewhat bigger than the nominal size, of course).

I think your hypothetical sleeve would be going the wrong way, though. To my mind, the "mortar" is cycling too fast for the front end "flask" - i.e. the flask needs to be speeded up. The way to do this is to decrease the impedance of the neck of the flask (the intake pipe), not increase it. Try this with NUDiS - set up a little flask of reasonable dimensions and a low pressure in the body of the flask, then play with the intake dimensions and you'll see what I mean. A bigger (or shorter) pipe increases frequency, because the lower impedance lets the mass "pour" in and out faster (meaning, higher massflows, NOT higher speeds) - that's really all there is to it.

This flame out during temperature increase is very very interesting to say the least, and you say that your intake temperature is perfectly stable, what if I put this to you...

Since the intake is drawing big breaths across that steadily increasing surface temp, might that have been something we completely brushed over but should have considered a little more closely. Is it possible its having a slightly increased effect having the intake so close, what do you think about that, I really can't be sure?
Anyway thats a bit of excited speculation from my part, just wondering how you felt about that idea?

That would be an interesting hypothesis, if it weren't for one little thing: I already have some experience with engines whose intakes lie right down on the tailpipe!!! The Fo Mi Chin II, the Sveldt Lady and the Slim Lady were all set up just like this (or even tighter!) and they were all stable running engines that would run forever once they were up to speed. Of course, in this case, the radiating area is bigger than a straight tailpipe, but I don't think that's enough difference to make your speculation viable. It's cool thinking, though, even if I'm not buying it ;-)

Good stuff larry, I bet you jumped for joy when she fired up that fast and easily! You know she is a real beauty, your skill on these matters has always been incredible, and I'm sure you'll have that whopping great blue aura out the tailpipe real soon, keep up the good work, and congrats.
Now, you'd better go catch your horses because at the frequency these things run at I bet they scattered like crazy :-)

Well, of course, I was expecting her to jump to life right away ;-) If it had acted like my earlier front-loaded Lady Anne attempts, I would have been completely surprised (and utterly disgusted) - the de-tuning over several seconds of run time was not amusing, but at least we must be pretty close. A few mm somewhere is going to bring it in, James - you can bank on it!

A comment might be in order here on construction accuracy. I have never produced an engine that was "right on the money" with the design dimensions. Jim and Jason did way better than I usually do - the only basic difference is that the chamber is about 3mm short, probably because of the necessity of removing and re-attaching the dome, and the engine as a whole is 4mm short. On the other hand, when I weld them up myself, I always end up a fraction of an inch OVER design length - and I mean, ALWAYS, although my steel version of this engine is only about 4mm too long. The layout of the intake on the stainless one appears to be spot on in terms of the chamber-to-mid-cone seam, which means the port is 3mm closer to the dome than designed, due to the shortening of the chamber just mentioned. I doubt that this slight shortening is the cause of the problem - it shortens BOTH the mortar and the flask, after all. I think the intake layout on my steel model is really close to the design dimensions, also - the smaller ID being the exception.

The only way the horses would run away is if I did this somewhere nearby. The more usual reaction (for Arabians, anyway) would be to look up, trot over to the near corner of the fence and stand there with ears perked up and noses pointed toward the source of the noise. It must be pretty frustrating to not be able to see what's producing that sound, with it sort of "coming out of nowhere" like space invaders or something.

L Cottrill

[larry cottrill]

Just realized I haven't mentioned yet how we came out for engine weight. With spark plug and fuel pipe & clamps in place, we're right at 20 ounces. That's more than I wanted, but part of it is our decision to use almost 1/16-inch stainless for the front dome. This is an attempt to keep the engine from bashing in the front dome under multiple G "crash loading" forces. It made sense to do this because of my intent to use the single-point thrust mount at the spark plug.

L Cottrill

[hagent]

Larry,

I'm glad that you are so close to your new design.

I'm sure you'll nail it down on your next run.


Your success make me want to go back to my concentric designs and impove upon them. I guess not before I get Eric's big one going well.

I'm happy to hear the good news.

Congrats...

[larry cottrill]

GOT IT! Both running out Saturday afternoon, but couldn't post anything because our home PC needed to be re-built (now done, mostly).

I decided that the very next thing to do was try the plain steel engine, to verifiy that it showed the same behaviour. The only change in procedure was to go with a higher fuel vapor pressure, 20 PSIG, since what I had done with the stainless Lady was obviously on the weak side. She started right up ... and just kept on going!

How could this be? The two motors are only different in two details: The intake on the stainless engine is a couple of mm larger ID; and the stainless engine is 7 or 8 mm shorter overall. Now, I know that everybody KNOWS that on small pulsejets, a little error can mean everything, but guys, this is ridiculous! That 7 or 8 mm would be about 1.5 PERCENT of the QUARTER WAVE LENGTH! And the smaller intake ID would make the intake somewhat higher Q, which should make the steel engine harder to run, NOT easier! All this made absolutely no sense to me.

Of course, then it hit me: The higher fuel pressure (and consequently higher flow) is the answer. I let the steel engine cool a while and went down to the basement and grabbed the stainless Lady. In a few minutes, I had her set up, with the fuel pipe moved over from one to the other. When I tried it, she started right up, and ran ... and ran ... and ran! That's all there was to it. The engine still seems to be throttleable, though it turns out this fuel setup, with low-pressure regulator and 3/16-inch lemniscate fuel pipe, is too restrictive - I can't get her up to rich extinction, and still no flame observed exiting the pipe. Unlike the earlier runs, the frequency went right up to a stable tone immediately on starting, and visible heating of the chamber was very rapid.

Man, this is a LOUD sucker - quite unpleasant, even with foam plugs in place AND hard shells on my ears. It must be the frequency. Amusingly, I noticed an occasional odd, brief "hollow ringing" sound that was quite disturbing - until I realized it was occurring when I happened to open my mouth slightly! I then tried "tuning" my mouth as a Helmholtz resonator, much as you do when singing. You should try that sometime with your small pulsejet running: When you get "in tune", the internal standing wave makes all your teeth "buzz" in a way that you wouldn't want to prolong if you have any loose fillings ;-) Sort of like having a dental drill working on all your teeth at once! Anyway, altogether lots of pulsejet fun for the half hour or so invested.

Another important observation with this design was that again, there was NO observable "overheating" of the chamber zone in the "tight spot" between the chamber and the intake pipe, either in the steel or stainless builds. As I imagined, there seems to be just enough heat sinking through the lateral weld between the two pieces (at the rear edge of the port) to compensate for the lack of good convection around that area of the chamber wall. If so, that detail has worked out amazingly well.

Heating patterns on these engines are particularly smooth, uniform heating throughout the chamber. Further aft, only the front end of the middle cone was really glowing - another indication that the fuel delivery is too restricted for full-power running. The tail cone, though not observably glowing, was heavily discolored in purples and browns. The chamber is now a dull desert brown color overall - as uniform as if it were spray painted, except for the front dome (which is heavier and gets some heat sinking from the thrust mount). The intake is unaffected; the transition fairing is a yellow color, like a thin flashing of gold over the steel. I'm talking about the stainless model, of course - the chamber of the steel version is dark grey and scaly, as is typical of cold-rolled steel engines after the initial run.

Of course, I have a couple of hypotheses as to why these engines are now working while exhibiting such odd behaviour before. The worst possibility is that the engines are only capable of running with fairly strong air entrainment from the action of fuel spouting - i.e. "pressure jetting". Another explanation might be that with the fuel pressure low (and therefore, spouting velocity inadequate), the engine was able to valve off (or perhaps blow back) fuel more effectively as the engine gained power during warmup until the engine reached a point of being "self defeating" - i.e. too lean to run. Another possibility is that the design IS slightly out of tune, but is most sensitive to the resonance error at very low power levels (which makes sort of intuitive sense to me).

At any rate, what we do know is: Anyone should be able to build one of these engines from my plan dimensions and have a running engine with propane vapor fueling.

L Cottrill

[hagent]

Larry thats Awesome!

I wonder how much the placement of your fuel injector will affect the running when using your low pressure propane setup?

Great job and can't wait to see some video and numbers.

[Irvine.J]

Excellent Larry!

"hollow ringing" sound that was quite disturbing"

oh yeah, makes you cringe doesn't it

Well done man, great to see this is going so well!

Of course, I have a couple of hypotheses as to why these engines
are now working while exhibiting such odd behaviour before. The worst possibility is that the engines are only capable of running with fairly strong air entrainment from the action of fuel spouting - i.e. "pressure jetting". Another explanation might be that with the fuel pressure low (and therefore, spouting velocity inadequate), the engine was able to valve off (or perhaps blow back) fuel more effectively as the engine gained power during warmup until the engine reached a point of being "self defeating" - i.e. too lean to run.

Luckily, I think its the later one, its not entrainment, its intake path is just really short and your really need good mixing AND pressure to make it run full hard. (in my opinion) Humor me and try a double D with an unregulated propane tank, its has a lower velocity compared to that 3/16 leminscate, and I think you'll find pressure/mixing and flow is the critical part. I like to think of it like a lady, (You have to know how to push her buttons:-) ) It will work and work damn well, in 1/4 inch, but possibly better in 3/16th as I've not tried it, you don't need braze it, just tap the "finger" down.

Fuel pressure is critical with this type as you say, the inlet blowback (relative to the engine size) is enormous, but its advantage is quite obvious, when its moving into the CC, it really slams that fuel/air mix back into the chamber violently!
Please fill me in on your observations with the intake's exhaust pattern, I have to say its extremely interesting and quite profound.

and the stainless engine is 7 or 8 mm shorter overall.

What!? Man, you might have stumbled onto one that runs even better!

Well, its certainly exciting times larry, your engine sounds like a real cracker, when you get the chance to take some video of it I'd love to see it, get her running on unregulated supply though I think your hair will be just that little bit more frizzy Maybe wear a mouthguard?

WAY TO GO!!!

Good stuff man, really excited for you, can't wait for further write-ups.
Cheers!

[larry cottrill]

Well, it's now sort of "ho hum" as Mark would say, but here's the video of the initial test runs where I couldn't get her to run out. Pretty draggy in spots, as I am not exactly a "Bob Vila of pulsejets". Some good examples of startup, though (incidentally, the starting is even better with higher fuel pressure - practically "instant on"):

http://www.youtube.com/watch?v=xKKipJwl2Mk

Enjoy!

I'll see what Jim and Jason are doing this week - Jim says he wants a demo right there in his artist studio! I'm just not sure when we'll get to it. But, I'll definitely get some video of that! In the meantime, I'll try to get a good run video back at the front garage doors of the Cottrill Cyclodyne labs. Stay tuned ...

L Cottrill

[Vermin]

Hey all
Just watched the vid..... Larry excelent start up and run at lower ranges.... it appears to be...could this be a mixture thing...such as the intake is not large enough to allow a proper fuel air combonation at higher ranges as in an internal conbustion engine where the mixture just gets to rich??
Only an observation...the original FWE is very adept with having two intakes.....curiosity only
V

[hagent]

Wow Larry,

Your lady there is running between 350 and 370 hz. You can hear it moving through cycles. It ran at 370 hz at the end of your video. I'm amazed at how high the frequency is.

I would be interested if you measured the frequency of the tail pipe in Nudis.

I just use a ruler on my monitor and count the wave form peaks with the time that you selected, and then do the math.

This is a good way (I think) to double check if Nudis is working well.

One of the pulse jets I built was dead on the money using this method, which really amazed me.

Let me know what you find.

EDIT -- OK I used the picture off your posted Nudis screen shot and the tail pipe is resonating at around 266 hz. You should try this your self because you can get a more accurate result than me. This would indicate a slight problem or inconsistancy with the actual run and Nudis. You may want to check your length measurements and adjust temp values to see if you can come close to the 350-370 hz. I think this will give you a more accurate nudis model. Just my 2 cents.

I'm going to listen to your test movie again and compare it at 266 hz just to make sure I wasn't matching up at a higher octive.

EDIT EDIT --- Ok it is definately not running at 266 hz. It's at 350-370 hz depending on how warm it is.

[hagent]

Hi Larry,

Just a correction. I ment Uflow instead of Nudis.

I put your Lady in Uflow and did some testing and I could not get the frequency up any where near 370 hz. I did get it up to 266 hz, but that is still a long way off.

I even closed off one end and did not even use the intake.

I'll have to try different things later.

Very interesting.

[larry cottrill]

Man - not even .003 sec per cycle. Amazing for a Dynajet-size beast like this.

I have never seen one of my UFLOW runs predict anything close to the actual cycling frequency. And, you're right - the error has to be in the internal temps assumed. However, all I care about is getting an engine that I know will resonate - this will predict success if all the temps are more-or-less equally underestimated, which mine always seem to be.

Similarly, I don't try to predict thrust with UFLOW. To compare two design variations, I compare the Mach values and Mass Flows, for both the intake and tail end. My thrust estimates for a design are just rule-of-thumb determinations, based on the smallest cross-section areas I want to use (or as modified by playing with the design in UFLOW). I don't pretend to be able to hit desired performance with great precision.

If we had a way of actually measuring temps accurately, I would certainly refine my UFLOW temps, and could probably bring the methods I use into pretty close alignment with reality. As it is, you could raise all the temps until you get the real frequency - but you'd still be guessing throughout the process. I'm just happy for us to come up with running engines.

L Cottrill

[hagent]

Hi Larry,

I was playing with an online resonance program.

http://hyperphysics.phy-astr.gsu.edu/hb ... ol.html#c1

And if it's correct your PJ is 0.546 meters long and it would appear that it's acting like a closed pipe resonator with an average temperature of 700-800 C. At 800 C it's right at 370 Hz.

So I believe that Uflow is not able to model it correctly as a closed tube.

What do you think?

I'm not an expert on this stuff.

Cheers,